import requests, os, sys, json, lzma
from bs4 import BeautifulSoup # pip install beautifulsoup4
import trafilatura # pip install trafilatura
from multiprocessing import Process
from multiprocessing.pool import Pool

stories = []
with lzma.open("3-stories.jsonl.xz", "rt") as fin:
    for idx, line in enumerate(fin):
        data = json.loads(line)
        print(data["title"])
        data["_id"] = idx
        stories.append(data)
print(len(stories))

#sys.exit()

def combine(story):
    _id = story["_id"]
    urls_contents = {}

    chapters_file = f"stories/{_id}_chapters.jsonl"
    if not os.path.exists(chapters_file): return _id

    story_file = f"stories/{_id}.jsonl"
    if os.path.exists(story_file):
        x = os.path.getsize(story_file) / os.path.getsize(chapters_file)
        if x > 0.8:
            try:
                with open(story_file, "rt") as fin:
                    for line in fin: 
                        json.loads(line)
                    return _id
            except:
                False

    with open(chapters_file, "rt") as f:
        for l in f:
            d = json.loads(l)
            urls_contents[d["url"]] = d["content"]

    title = story["title"]; cat = story["cat"]
    text  = f"Truyện: {title}\nThể loại: {cat}"
    count = 0

    chapter_urls = set()
    for chapter in story["chapters"]:
        t = chapter["title"]
        u = chapter["url"]
        if u not in chapter_urls:
            chapter_urls.add(u)
            if u in urls_contents:
                count += 1
                c = urls_contents[u]
                text += f"""

{t}

{c}

"""
        if count > (2*len(chapter_urls)) // 3:
            with open(story_file, "wt") as fout:
                fout.write(json.dumps({
                    "text": text,
                    "url": story["url"]
                }, ensure_ascii=False) + "\n")

    return _id

with Pool() as pool:
    count = 0
    for story_id in pool.imap_unordered(combine, stories):
        count += 1
        print(count, story_id, "DONE!")